Acute leukemias, the most common malignancy of childhood, have long been suspected of having an infectious etiology. We hypothesize that the common species C adenoviruses are responsible for the initiating step in some acute childhood leukemias. This hypothesis is based on the observation that species C adenovirus DNA is more frequently found in the blood of newborn children that later develop acute lymphoblastic leukemia than in the blood of children that do not develop leukemia. Additional support derives from observations made in the Ornelles and Gooding laboratories indicating that adenovirus infection of a lymphoid progenitor cell could produce the expanded clones of translocation-bearing cells that are the starting point for leukemia development in children. Studies proposed in this application will test this hypothesis through the following specific aims: (1) To evaluate the coincidence between prenatal infection with species C adenovirus, the occurrence of leukemia-associated translocations, and the eventual development of leukemia using archived Guthrie cards and fresh cord blood. (2) To elucidate the impact of adenovirus on DNA-repair in lymphoid cells. (3) To determine the impact of adenovirus infection on the integrity of lymphocyte cell DNA. (4) To evaluate adenovirus replication and the cytopathology of adenovirus in leukemic cells and hematopoietic progenitor cells. The studies proposed in this application will test directly the possibility that species C adenovirus infection is one step in the sequence of events leading to childhood leukemia. In addition, this work will identify cellular genes that affect adenovirus replication in lymphoid cells and determine the frequency of prenatal infection with this virus. These studies may provide support for hit-and-run mechanisms, that have been postulated for several viruses, in human oncogenesis.